concept
active
concept:emotional-state-persistenceEmotional State Persistence
The property of emotion features maintaining elevated activation well beyond the local token context that triggered them
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- Measures emotion feature persistence as correlation between z-scored activation at token 0 and token 100 across all eligible target model tokens
Concepts (1)
concept
- Stateful Internal Representationassociated_withA representation that maintains stable activation across many tokens rather than being locally triggered by specific content
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The phenomenon that emotion feature activations remain elevated above baseline beyond local token bursts, measurable as long-range correlation
- The key-value cache from steered tokens is retained during no-steering continuation, allowing causal effect of steering to propagate
- The central research question motivating the paper
- Psychological states like emotions, moods, pains, itches, defined by valence and arousal.
- The phenomenon where activating an emotion feature leads to subsequent below-baseline activation of that feature
- The possibility of a stably encoded, causally active emotional state within LLMs, as distinct from token-by-token semantic content
- Core unresolved confound the paper acknowledges but cannot rule out
- Proposed mechanistic explanation for why emotion features are more persistent